Page 1
Soil & Tillage Research 70 (2003) 1–18
Review
Is there a critical level of organic matter in the
agricultural soils of temperate regions: a review
P. Loveland
a
, J. Webb
b
,
∗
a
National Soil Resources Institute, Cranfield University, Silsoe, Bedfordshire MK45 4DT, UK
b
ADAS Consulting Ltd., Woodthorne, Wergs Road, Wolverhampton WV6 8TQ, UK
Received 6 August 2001; received in revised form 16 September 2002; accepted 16 September 2002
Abstract
Soil organic matter (SOM) is a complex mixture, which influences a number of soil properties and nutrient cycling, and is
itself influenced in kind and amount by land-use, soil type, climate and vegetation. There is considerable concern that, if SOM
concentrations in soils are allowed to decrease too much, then the productive capacity of agriculture will be compromised by
deterioration in soil physical properties and by impairment of soil nutrient cycling mechanisms. This has clear implications
for the sustainable use of soil. We have focussed our discussion from the standpoint of the sustainability of UK agriculture,
because we know that best, but similar concerns are equally valid elsewhere in the world. Although soil scientists would expect
to find different behaviour in different soils at different ‘critical’ concentrations of SOM, it seems widely believed that a major
threshold is 2% soil organic carbon (SOC) (ca. 3.4% SOM), below which potentially serious decline in soil quality will occur.
This review summarises what is known about critical thresholds of SOC or SOM, mainly in soils of temperate regions.
It examines critically the quantitative, rather than anecdotal or descriptive, evidence for such thresholds and their potential
effects on soil quality, soil physical properties and crop nutrition, and the links between these.
We conclude that the quantitative evidence for such thresholds is slight, although there is some evidence that there might be
an desirable range of SOC covering a wide spectrum of soils, but again the quantitative evidence for this needs considerable
development. There is also little quantitative evidence that reduction in SOC concentrations in the soils of England and Wales
will have marked effects on other soil properties or crop yields. The data do suggest, however, that more research is required
on the nature of SOC, particularly of the so-called ‘active’ or ‘fresh’ fraction and its influence on the properties of a range of
soil types under different land uses. This is particularly relevant to the ongoing debate about soil quality, its definition, and
the identification of appropriate indicators that relate soil quality to soil functions.
© 2002 Elsevier Science B.V. All rights reserved.
Keywords: Soil organic carbon; Soil organic matter; Critical levels; Soil quality; Temperate soils; England and Wales
1. Introduction
It is well-known that additions of organic matter
(OM), e.g. manures, composts, above-ground crop
∗
Corresponding author. Tel.:
+
44-1902-693235;
fax:
+
44-1902-693166.
E-mail address: m99102@adas.co.uk (J. Webb).
residues, or increases in soil organic matter (SOM),
e.g. below-ground crop residues, microbial biomass,
can improve soil properties. This was, and contin-
ues to be, the basis of much traditional farming,
where large amounts of mineral fertilisers are either
not available or are eschewed, or where tillage/short
grass-ley rotations (with or without animal husbandry
as part of the cycle) are preferred over long-term
0167-1987/02/$ – see front matter © 2002 Elsevier Science B.V. All rights reserved.
PII: S 0 1 6 7 - 1 9 8 7 ( 0 2 ) 0 0 1 3 9 - 3
Page 2
2
P. Loveland, J. Webb / Soil & Tillage Research 70 (2003) 1–18
monoculture. In recent years, organic farming has in-
creased, wherein a move from industrially produced
fertilisers to ‘natural’ products is a central part of
the farming system (see e.g.
Lampkin and Spedding,
1990
). An increase in SOM is also still seen, by
many conventional farmers, as a desirable objective.
Better plant nutrition (N, P, S, micronutrients), ease
of cultivation, penetration and seedbed preparation,
greater aggregate stability, reduced bulk density (D
b
),
improved water holding capacity at low suctions,
enhanced porosity and earlier warming in Spring
have all been associated with increased amounts of
SOM (
Carter and Stewart, 1996
). It follows from this
that a decrease in SOM, e.g. by oxidation following
cultivation, especially under long-term tillage, the
withdrawal of grass-leys from rotations, the absence
of animal or ‘green’ manuring, or long-term climate
change, might adversely affect most, if not all, of
these properties. This decline might occur even if,
in the long term, SOM reaches an equilibrium with
additions of OM in crop residues (
Arrouays and
Pelissier, 1994
), composts, sewage sludge or other
organic wastes (
Mamo et al., 1999
). There are clear
relationships between these different perceptions of
SOM and current views on definitions of soil quality
(‘soil health’), and the identification and develop-
ment of appropriate soil quality indicators and soil
monitoring programmes (see e.g.
Davidson, 2000
).
Kemper and Koch (1966)
indicated that many soils
in the western United States and Canada suffered sig-
nificant decline in structural stability if soil organic
carbon (SOC) less than 2%. Similarly,
Greenland et al.
(1975)
concluded that soils in England and Wales
with
<
2% SOC were prone to structural deterioration.
However, this 2% threshold was a ‘rule of thumb’
(sic) to indicate soil structural stability, not a measure
of soil physical properties in the field. Nevertheless,
this value of 2% SOC (equivalent to ca. 3.4% SOM
1
)
1
‘Humus’, ‘(soil) organic matter’ and ‘(soil) organic carbon’
are still used interchangeably, often within the same document.
We have used the terminology of original papers in what appears
to us to be the most consistent manner. It is usually soil organic
carbon which is measured quantitatively, and we have used the
convention that 1% SOC
=
1
.
7% SOM, although this is by no
means the constant it is often said to be (
Howard and Howard,
1990
). Measurements of soil organic matter by ‘loss-on-ignition’,
i.e. low-temperature combustion, are even more variable, and their
conversion to soil organic carbon should be made with caution.
is often taken by those with concerns over the im-
pacts of intensive farming practices as a threshold be-
low which the soil becomes physically unstable, more
susceptible to cultivation damage and to erosion, and
that society should be concerned when a significant
proportion of soils are below this value (e.g.
Pretty,
1998
). Crop yields might suffer due to reduction in
the soil’s capacity to cycle nutrients. This will, in turn,
lead to reduction in the return of SOM to the soil via
crop residues and the cycle of deterioration will inten-
sify. The Royal Commission on Environmental Pollu-
tion (RCEP), in its report on ‘Sustainable Use of Soil’
(
Department of the Environment, 1997
), commented
on the undesirability of allowing SOM to decrease too
much, although it avoided recommendation of limit-
ing values. Thus, a concept of a critical threshold for
SOC has become widespread and bought to the atten-
tion of policy makers who have sought clarification of
the scientific validity of this concept. Whilst soil sci-
entists might look askance at the suggestion of a single
threshold applicable to all soils and land uses, there is
clearly a need to review the position, summarise the
quantitative evidence for and against this, and promote
the findings to a non-specialist audience. This review
was commissioned by the then UK Ministry of Agri-
culture, Fisheries and Food (MAFF) to address these
concerns and to assess whether 2% SOC is a critical
threshold of soil quality.
Consideration of such critical levels involves assess-
ment of the quantitative evidence, i.e. the nature of
SOM and the properties it confers on soils, whether
justifiable limits can be set for a range of soil types, cli-
matic conditions, or land management/cropping prac-
tices and, finally, whether there are any dis-benefits
from an increase in SOM levels in soils. In other
words, are there critical upper levels of SOM? We em-
phasise, again, that we were considering the quantita-
tive evidence; we have looked at much anecdotal and
descriptive reporting, but these sources of information
are rarely accompanied by quantitative data.
2. SOM—some general considerations
A detailed discussion of the physico-chemical na-
ture of SOM and related factors, for which there is
a very large literature (see e.g.
Hayes and Wilson,
1997
), is beyond the scope of this review. However,
Page 3
P. Loveland, J. Webb / Soil & Tillage Research 70 (2003) 1–18
3
certain points are germane, and are discussed below.
For many hundreds of years, much farming practice
was based on the sound observation that additions of
relatively fresh organic materials to soils improved
tilth and crop yields, i.e. they affected soil physical
properties and crop nutrition. This perception of the
benefits of ‘fresh’ SOM persists to this day (see e.g.
Watts et al., 2001
). Animal manures, composts, crop
residues, sewage sludge, etc. are seen as inherently
‘better’ for soils than ‘old’ or ‘inert’ forms of OM,
such as wood products (
Lampkin and Spedding, 1990
),
lignite (
Fagbenro and Agboola, 1999
) or coal dust
(
Zier et al., 1999
).
This separation of SOM into two broad classes,
namely ‘active’ and ‘inert’ is widely reported in the re-
search literature (e.g.
Tisdall and Oades, 1982; Carter
and Stewart, 1996
), but less so in much of the applied
literature, where total SOM is usually measured and
discussed (e.g.
MAFF, 1998
). Total SOM is conve-
niently measured as SOC, e.g. by methods based on
oxidation by acid dichromate solution (
Kalembasa and
Jenkinson, 1973
), or by ‘loss-on-ignition’ (
Howard
and Howard, 1990
). Measurements of ‘active’ SOM
often focus on both detailed investigations of particu-
lar compounds or classes of compounds (see e.g.
Beck
et al., 1993
), or on relatively simple determinations
of the amounts of broad classes of compounds in the
soil, e.g. of water-soluble mono- or poly-saccharides
(
Martens and Frankenberger, 1992
), ‘gums’ (
Chesters
et al., 1957
), easily extracted nitrogenous species such
as peptides and aminoacids (
Tisdall and Oades, 1982
),
or as carbon readily oxidised by cold or weak solu-
Table 1
Mean and range of soil organic carbon content (%, in
<
2 mm air-dried soil) under different land uses and for different soil texture groups
(
Clayden and Hollis, 1984
) based on National Soil Inventory data for England and Wales (Bellamy, pers. comm.)
Soil texture group
Soil organic carbon (%)
Arable
Ley grass
Permanent grass
a
n
Mean
Range
n
Mean
Range
n
Mean
Range
Clay
290
3.38
1.2–9.9
72
4.3
1.3–10.9
208
5.45
1.3–10.9
Coarse loamy
430
1.97
0.5–8.1
125
3.05
0.9–9.0
281
4.06
1.1–11.9
Fine loamy
651
2.60
0.5–10.9
306
3.51
1.0–12.0
631
4.34
1.1–11.5
Coarse silty
22
2.85
1.1–5.9
10
3.17
0.8–6.6
18
3.86
1.5–10.6
Fine silty
186
2.72
0.8–10.1
88
3.44
1.0–9.6
166
4.36
1.0–11.0
Sandy
75
1.75
0.4–9.1
7
2.29
0.7–4.5
28
2.59
0.6–5.4
Peat
54
19.75
12.1–56.4
12
20.98
12.2–49.2
54
20.61
12.2–52.7
a
Managed grassland only, excludes rough grazing and open moorland.
tions of an oxidant such as KMnO
4
(
Bell et al., 1999
)
without further differentiation of the substances ex-
tracted. Some active components have been associated
with fungal hyphae (
Beare et al., 1997
), others with
root functions (
Miller and Jastrow, 1990
) and also with
microbial activity (
Tisdall, 1994
).
Differentiation of SOM into ‘active’ and ‘inert’
fractions is important in terms of SOM turnover. In
general, the easily oxidisable components of SOM
have half-lives
<
1 year, whereas forms associated
with longer term accumulation of SOM have half-lives
up to 25 years or more (
Tisdall and Oades, 1982
). The
ratio between short-lived SOM and long-term SOM
is often assumed to be relatively constant once the
SOM is at or close to equilibrium, and this assump-
tion is built into many mathematical models of SOM
turnover (
Smith et al., 1996
). However, it is important
to realise that establishment of such an equilibrium
can take decades, and the process can be subject to
considerable short-term disturbance by agriculture.
In areal terms, there are ca. 6
×
10
6
ha of land
in England and Wales under arable cultivation, and
5
.
5
×
10
6
ha under permanent (managed) grass >5
years (June 1999 data; MAFF Statistics). There is
a further 5
.
7
×
10
6
ha described as rough grazing.
These broad divisions of land-use are crucial in terms
of additions of SOM, especially of ‘active’ fractions,
wherever they occur in the world. Grass produces
crop residues and root exudates almost continuously,
whereas tilled crops go through an annual cycle of
build-up of ‘active’ fractions of SOM, which then de-
cline as the short half-life components turnover in the
Page 4
4
P. Loveland, J. Webb / Soil & Tillage Research 70 (2003) 1–18
soil. Further, the breakdown of SOM residues in soils
tends to be faster where it is warmer. Large areas of
soil in the warmer east and south of England, where
arable cultivation dominates, have an SOC content be-
low or close to 2% in the cultivated horizon. Amounts
of SOC are generally greater in the cooler, wetter re-
gions of Wales, south-west England and northern Eng-
land, where long-term grassland dominates (
McGrath
and Loveland, 1992
). Similar geo-climatic factors can
also be seen for France (
Arrouays et al., 2001
). Fur-
ther, soil type also affects SOC content; the greater
the clay content of the soil, then the greater is SOC,
assuming similar land-use and management practices
(
Table 1
).
Against this background, we examined the literature
on the contribution of SOM (or SOC) to soil prop-
erties, concentrating on quantitative evidence, mostly
on aspects relevant to UK agriculture, and we regard
this as an adequate discussion-model for much of the
world’s temperate agriculture.
3. Influence of SOM on crop production
Allison (1973)
summarised the role of SOM in crop
production under four headings: crop nutrition, mi-
cronutrient availability, soil available water, and soil
buffering capacity, viz:
1. The ‘active’ pool of SOM, i.e. plant and animal
residues added to the soil in the previous 5, or
at most, 25 years, provides N for crop growth,
and increases the availability to plants of many es-
sential micronutrients. Older SOM makes almost
no direct contribution to crop nutrition. This con-
cept of ‘young’ OM has been developed by many
later workers, e.g.
Janzen (1987)
and
Matus and
Rodr´ıguez (1994)
.
2. SOM has little direct effect on soil available water,
or indirect effects via changes in pore size distri-
bution, except in sandy soils, largely because it is
held at too high a tension to be available to plants.
3. Chelating and buffering are considered by many to
be the most important properties of soils, as without
these agriculture would require constant manage-
ment. CEC is typically 2–50 cmol
c
kg
−
1
in min-
eral soils, with values
<
3–4 cmol
c
kg
−
1
in sandy
soils, and SOM may account for 30–60% or more
of this. Soil CEC may increase with increasing pH,
and by addition of phosphate (
Tan and Dowling,
1984
). Hence CEC comprises a permanent (CEC
p
)
and variable (pH dependent) fraction (CEC
v
). The
CEC
p
has been regarded as derived mainly from
the clay fraction, while CEC
v
is mainly located in
the soil SOC.
This review of subsequent work on the effect of
SOM on crop production follows these same headings.
3.1. Crop nutrition
The role of SOC may be estimated either from com-
parison of soils of the same type, but of different SOC
status, or from long-term studies in which crop yields
are related to changes in SOC over time. Unambigu-
ous identification of SOC effects is difficult due to
confounding of changes in SOC with changes in other
factors, although comparison of soils of differing de-
grees of degradation are usually less equivocal.
Battiston et al. (1987)
studied erosion phases de-
veloped on the same parent material within fields
managed as a single unit since cultivation began. In
severely eroded soils, SOC was up to 56% less than
in uneroded soils, but not consistently so. The main
causes of yield reduction of grain maize (Zea mays L.)
were: reduced seedling population and vigour due to
poor seedbed conditions (exposure of clay subsoils);
nutrient deficiencies (doubtful from values quoted);
moisture stress (reduced profile available water
capacity—AWC) by exposure of sandy or imperme-
able subsoils).
Verity and Anderson (1990)
found that
topsoil erosion reduced both SOC from ca. 4.5–0.5%,
and also total and mineralised N and S. Spring wheat
(Triticum aestivum L.) yields were greater at the foot
of slopes than on the eroded parts. There was no
assessment of fertiliser additions in ameliorating the
effects of soil erosion.
Tanaka and Aase (1989)
mim-
icked erosion by systematically removing topsoil, viz:
control; 50% of A horizon removed; all of A horizon
removed; all of A horizon and some subsoil removed
(mean SOC in the top 30 cm: 1.27, 1.18, 0.98 and
0.96%, respectively). In three out of 5 years, yields
of Spring wheat (T. aestivum L.) were reduced by
topsoil removal, but addition of N and P fertilisers
brought yields from treated soils up to those of the
control. In the other 2 years, yields were limited only
Page 5
P. Loveland, J. Webb / Soil & Tillage Research 70 (2003) 1–18
5
by water supply. In all these experiments, P was the
limiting nutrient.
Robbins et al. (1997)
also removed
30 cm of topsoil from a silt soil (SOC
=
0
.
94%, top-
soil; SOC
=
0
.
45%, subsoil), fairly uniform to 1.3 m
depth, to study subsoil amelioration after applying
adequate fertiliser N, P and Zn. Dry edible bean
(Phaseolus vulgaris L., cv. Viva) yield was less on
the subsoil than the topsoil when fertiliser alone was
applied, but equivalent yields were obtained when
the subsoil was amended with cattle manure. Plant
analyses indicated that Zn may have been limiting
except for the topsoil and manure-amended subsoil.
The authors concluded it was not possible to deter-
mine whether yield increases were due to the Zn,
or SOC supplied by the manure, or by both. Simi-
larly,
Zhukov et al. (1993)
found that, for a loamy
sand podzol, unfertilised grain yield (winter wheat
[T. aestivum L.], winter barley [Hordeum vulgare
L.], and winter rye [Secale cereale L.]) was directly
dependent upon ‘humus’ (total SOM) concentration.
Fertiliser-N largely nullified the effect of humus, but
even with application of 100 kg N ha
−
1
and 20 t ha
−
1
cattle manure, yields at 0.7% humus remained less
than those of the other humus treatments (96 and 85%
of maximum yields for wheat and barley, respectively).
Bauer and Black (1994)
quantified the effect of
varying SOC concentrations on spring wheat produc-
tion, using the upper horizon of a single soil type with
uniform available soil mineral-N (SMN) and avail-
able water. Fertiliser-N inputs were adjusted, against
measured SMN, to supply 55, 90 and 125 kg ha
−
1
available-N. In 1984 (SOC
=
1
.
28–1
.
69%) and 1987,
no differences in yield could be related to soil SOC
content. In 1985/1986 (SOC
=
0
.
97–2
.
51%), even
with maximum fertiliser-N and irrigation, yields on
the plots with less SOC were
<
ca. 90% of the max-
imum. This was considered to be due to the smaller
minimum SOC in those years.
Shepherd et al. (1996)
also observed that even on UK soils receiving large
annual fertiliser-N applications, a significant propor-
tion of crop N uptake was mineralised from SOM.
Johnston (1991)
, based on the long-term experi-
ments at Rothamsted, considered that for any one soil
it is difficult to determine a critical ‘humus’ content
below which yields would be adversely affected. The
Woburn experiment (sandy loam soil—Dystri-Luvic
Arenosol (
FAO, 1998
)) was established in 1937, on a
site that had been in an arable rotation since 1876 and
had only ca. 1% SOC. Despite a continued decrease in
SOC over the next 30 years by ca. 0.5% there was no
catastrophic reduction in yields. In fact, provided fer-
tiliser applications, particularly N, were appropriate,
and the build-up of soil pathogens was avoided, yields
tended to increase. In contrast, the SOC content of the
unmanured plot in the Hoosfield continuous barley ex-
periment at Rothamsted has remained unchanged at ca.
1.2% for 100 years (
Jenkinson and Rayner, 1977
). The
unmanured plot at Broadbalk (Profundi-Endostagnic
Luvisol;
FAO, 1998
) also showed no decrease in SOC.
Some of the other plots have been kept free of plants
since 1870, and SOC has decreased further. This sug-
gests that, in the cropped fields, SOC is maintained
by crop-C inputs. The addition of 7.5 t ha
−
1
straw
for 6 years to the sandy soil at Woburn increased
SOC from 0.7 to 0.9% (
Johnston, 1991
). This increase
in SOC was associated with consistent increases of
arable crop (winter wheat, barley, potatoes (Solanum
tuberosum L.) and sugarbeet (Beta vulgaris L.) yields
over a range of fertiliser-N applications. Since the
mid-1960s, evidence has accumulated from plots given
an annual application of farmyard manure (FYM) of
an additional benefit to crop yields from this even with
fertiliser application. Before then, yields with fertilis-
ers alone were equal to, or greater, than those given
FYM.
Johnston (1991)
related these benefits to the
introduction of cultivars with greater yield potential,
together with agrochemicals to control pests, weeds
and disease. The annual application of FYM increased
SOC to ca. 3.4% compared with ca. 1.0% initially,
but since fresh FYM was applied each year any crop
response to increased SOC only could not be identi-
fied. The FYM supplied the equivalent of an additional
69 kg ha
−
1
fertiliser-N applied in the Spring, which
gave a yield increase of 1.4 t ha
−
1
over fertiliser-N
alone. Thus annual addition of organic manures may
increase yields, even with the application of large
amounts of fertiliser. However, it is unclear whether
the benefit is due to the FYM or to increased SOC per
se, or to an increase in some related property such as
soil available water content or minor nutrients in the
manure. Broadbalk field now has SOC concentrations
of only 0.9–1.0% (
Glendining et al., 1996
). Yet, where
192 kg ha
−
1
fertiliser-N is applied to winter cereals
(the average fertiliser-N application when the SMN is
small (N index 0 for the UK)) (
Anon., 1994
), yields are
6–7 t ha
−
1
, average for the UK. While the long-term
Page 6
6
P. Loveland, J. Webb / Soil & Tillage Research 70 (2003) 1–18
application of fertiliser-N has had no measurable ef-
fect on soil SOC, total soil N is greater where more
fertiliser-N has been applied. There has also been a
proportionately greater increase in N mineralization.
Grace et al. (1995)
reported an overall decrease in
SOC from 2.75 to 1.56% in a long-term experiment
(begun 1925) in Australia, the decrease being greater
in rotations with a greater proportion of fallow. Yields
have tended to decrease with time. Overall, the sec-
ond largest yields (2.21 t ha
−
1
) were produced by a
wheat–pasture–fallow rotation, which experienced one
of the greatest reductions in topsoil SOC (60%), sug-
gesting a large release of nutrients (especially N) from
the SOM pool, the only fertiliser input being P. Thus,
in an unfertilised soil, in which breakdown of SOM
is necessary to maintain yields, there may be a criti-
cal SOC below which insufficient nutrients are miner-
alised to sustain satisfactory yields. However, since the
nadir of yields in the 1960s on the continuous wheat
plot there has been a subsequent increase in yields,
which the authors attribute to a gradual build-up of
light-fraction SOM, resulting from the greater C/N ra-
tio of cereal residues. This light fraction is the more
important in respect of aggregate formation and as a
source of nutrients (cf.
Allison, 1973
). Thus, from this
study, and those at Rothamsted, there is evidence that,
for cereal rotations at least, reductions in light-fraction
SOC can be arrested as returns from crop residues be-
come equal to the annual breakdown of SOC.
Decrease in total SOC may not be in proportion to
decrease in the release of nutrients.
Wadman and De
Haan (1997)
concluded, from a 20-year pot study of 36
soils, that those with small SOC contained relatively
large proportions of ‘young’ SOC, leading to greater
initial mineralization, and that the rate of SOC de-
composition decreased with time, eventually tending
toward zero (see also,
Arrouays and Pelissier, 1994
).
This is consistent with the results of many other work-
ers that it is the ‘active’ fraction that is most important
in nutrient supply, and that as long as crops are grown,
returns of OM in crop residues and root exudates are
likely to balance the annual breakdown of SOC.
Körschens et al. (1998)
present the results of some
long-term experiments (begun 1902), which included
treatments to ‘discern the influence of highly differ-
ent SOM contents on yield and C and N dynamics’.
SOM was regarded as comprising at least two frac-
tions, one largely inert, the other mineralizable (C
min
).
Only C
min
is considered important. When averaged
across all crops and years, no positive yield response
was recorded, either from greater SOC, or FYM ap-
plication when mineral fertilisation was optimal. The
authors concluded that, in almost all long-term experi-
ments, C and N concentrations in soil have reached dif-
ferent equilibria. For a sandy soil containing 3% clay,
SOC was 0.7%, while for a loess soil with 21% clay,
SOC was 2.0%. For both soils, C
min
was 0.4–0.5%.
The contribution of SOC to crop yield was estimated at
no more than 10%, by which the authors mean that at
least 90% of potential yield can be obtained on sandy
soils with addition of mineral fertilisers, and 95% of
potential yield on other soil types. The authors pro-
posed an upper limit to SOC, above which there is
an increased risk of pollution due to losses of N and
C (CO
2
). Lower limits were also proposed for total
SOC to maintain optimum crop production, which in-
creased with increasing clay content, 1% SOC at 4%
clay to 3.5% SOC at 38% clay. The upper limits pro-
posed were 1.5–4.4% SOC, respectively. The range is
equivalent to 0.2–0.6% C
min
.
The studies reported above cover a wide range of
soil types, crops and crop managements in a num-
ber of regions. Overall, the results of these studies
suggest very tentatively (this cannot be too strongly
emphasised) that, irrespective of soil type, if SOC
decreases to ca. 1% it may not be possible to ob-
tain potential yields, most probably due to reduc-
tions in mineralizable-N, even with addition of N,
or N and P fertilisers. However, in many studies
quoted additions of N and P were only moderate (ca.
40–125 kg nutrient ha
−
1
), and no S fertilisers were
applied. Hence, nutrient supply might have remained
limited, especially in relation to S and, perhaps, minor
elements (see also below). It is also possible that the
inability of fertiliser-N and P to fully maintain crop
yields was in consequence of better buffering of nutri-
ent supply in soils with greater SOC. In some of these
studies, satisfactory crop yields were, nevertheless,
obtained from soils of SOC concentration
2%. We
conclude that data from agricultural systems which
require the mineralization of SOM to maintain crop
yields, and in which decreases of SOC below a certain
level may lead to insufficient nutrient release, have
been erroneously applied to temperate agricultural
systems in which adequate nutrient supply is obtained
from fertilisers. If the concept of a critical threshold
Page 7
P. Loveland, J. Webb / Soil & Tillage Research 70 (2003) 1–18
7
or thresholds of SOC is to be sustained, in the context
of crop yields, then considerably better evidence is
required to support it.
3.2. Available water capacity
Bauer and Black (1992)
found that AWC remained
essentially constant in sandy soils as SOC increased
from 0.74 to 1.49%, while for the medium and
fine-textured soil groups, AWC tended to decrease
with increasing SOC. This was in consequence of bulk
density (D
b
) decreasing, pore size increasing and the
proportion of air-filled pore space increasing as SOC
increased in all three texture groups. They concluded
that the decline in productivity from soil erosion was
not caused by a reduction in AWC, but by a decline
in nutrients and biological activity.
Haynes and Naidu
(1998)
also concluded from a range of data that, since
the water holding capacity at both field capacity and
wilting point is generally increased by additions of
OM, the AWC is not greatly changed.
Thomasson
and Carter (1989)
showed that SOC contributes no
more than about 15% to the variance in soil water
content held in agricultural topsoils. Examination of
these data shows no threshold values below or above
which soil water holding capacity changes markedly
with SOC content.
3.3. Micronutrient availability
McGrath et al. (1988)
added solutions containing a
range of concentrations of Mn, Zn and Cu to five sam-
ples of sandy loam soil with SOC concentrations rang-
ing from 0.46 to 1.39%. After a 3-week incubation,
the total concentrations of Mn, Zn and Cu in displaced
solutions were almost independent of SOC, although
the concentration of Cu
2
+
was smallest with great-
est SOC. When EDTA and acetic acids were used as
single extractants, metal concentrations of samples to
which no metals had been added, were unaffected by
soil SOC. However, where Zn and Cu had been added
to the soils, acetic acid-soluble metals decreased as
SOC increased at all pH’s. However, this study was
of the effects on metals added to soil rather than on
availability of trace metals already present.
Tagwira
et al. (1992)
found that, on sandy soils of low Zn sta-
tus, addition of SOC may reduce extractable Zn avail-
ability at least in the short term.
Figliola et al. (1993)
added Cu and Zn to soils of contrasting SOC status.
Extractable Zn increased with increasing SOC, but this
was not accompanied by an increase in plant (Avena
sativa) uptake of Zn. Extractable Cu was reduced by
increasing SOC, but there was no effect on plant up-
take. Thus we conclude there is no consistent effect
of SOC on crop uptake of Zn or Cu.
3.4. Cation exchange capacity
Addiscott (1970)
found that ‘organic CEC’ was ca.
230
±
47 cmol
c
kg
−
1
for calcareous arable soils, whose
SOC contents ranged from ca. 0.8 to 3.2%.
Curtin
and Smillie (1976)
studied 51 soils, with mean SOC
of 1.9% (range 0.1–8.9%), mean clay 26.2% (range
0.4–56%) and pH 6.2 (range 3.8–8.3), and concluded
that CEC was correlated with SOC (
R
2
=
50%) and
specific surface area (
R
2
=
66%).
2
Multiple regres-
sion on both accounted for 97% of the variation in
CEC.
Martel et al. (1978)
in a study of 11 clay soils
(40–51% clay, mean 3.1% SOC) concluded that only
ca. 10–15% total CEC was associated with SOC. In
these soils too, surface area was better related to CEC
than clay content, albeit on these clay-rich soils CEC
was dominated by the mineral fraction.
Tan and Dowling (1984)
investigated the role of
SOC in variable CEC (CEC
v
) and permanent CEC
(CEC
p
) by destroying soil SOC with H
2
O
2
. Of the
five soils studied, total CEC was greatest in a soil with
montmorillonitic clay. In this soil, removal of SOC
increased CEC
p
, due to unblocking of exchange sites
by SOC. In other soils, reduction in SOC decreased
CEC
p
because of the reduction in CEC
p
in SOC. Vari-
able CEC was >40% of the total in all soils and ca.
50% or more in four soils. However, the study did not
confirm SOC as the major site of CEC
v
. These work-
ers concluded that the interaction between inorganic
and organic components, in providing sites for CEC,
both variable and permanent, was largely controlled
by the mineralogy of the soil.
McGrath et al. (1988)
reported CEC of a sandy soil to increase from 75 to
2
Use of R
2
(co-efficient of determination) and r (correlation
co-efficient): we have given whatever was quoted in the original
paper. We note that R
2
(a test of goodness-of-fit, applicable to
any kind of distribution) is now common in more recent literature,
whilst the use of r has declined. Note also that r can take a
negative sign, indicative of an inverse relationship, while R
2
does
not conventionally take such a sign.
Page 8
8
P. Loveland, J. Webb / Soil & Tillage Research 70 (2003) 1–18
158 cmol
c
kg
−
1
as soil SOC increased from 0.46 to
1.39%.
Thompson et al. (1989)
estimated the contribution
of SOC to soil CEC by means of an orthogonal com-
parison of soils, one cultivated after prairie, another
cultivated after deciduous woodland. Each was paired
with an uncultivated prairie or deciduous woodland
soil. Soils were of a common parent material and land-
scape position to standardise the effect of mineralogy
on CEC and specific surface area. Soil SOC was es-
timated to be responsible for ca. 50% of total CEC.
However, in the fine clay fraction, in contrast to the
coarse and fine silt and coarse clay fractions, inorganic
soil colloids dominated CEC. This was considered due
to organic colloids coating inorganic surfaces, hold-
ing particles together, and making net CEC less than
would be expected from an additive estimate.
Nearly all the workers quoted have discussed only
total SOC or SOM.
Haynes and Naidu (1998)
quote re-
sults from the long-term experiment at Askov in Den-
mark. Here the CEC of SOM derived via crop residues
from mineral fertiliser applications (560 cmol
c
kg
−
1
SOM) was greater than that derived from additions of
FYM (381 cmol
c
kg
−
1
SOM). It was concluded that
the SOM formed from crop residues had a greater pro-
portion of aromatic molecules than that formed from
FYM addition.
Thus, from these studies, it may be concluded that in
agricultural soils, maintained at pH of 6–8 ca. 40–50%
of CEC may be associated with SOC. However, the
proportion is likely to be much less in clay soils. There
is also an interaction with clay mineralogy. Soils con-
taining montmorillonite clays may have their potential
CEC reduced by SOC reducing surface area.
4. Soil physical properties
The addition of OM is considered to increase the
stability of soil aggregates, reduce D
b
, alter pore size
distribution leading to greater water holding capac-
ity, and reduce susceptibility to erosion (
Carter and
Stewart, 1996
).
4.1. Aggregate stability
There is little uniformity in the choice of meth-
ods for determining aggregate stability, including sam-
ple pre-treatment, e.g. standardisation of water con-
tent prior to measurement. Equally, there is no gen-
eral agreement as to the size range or size intervals of
aggregates whose stability is being reported. This can
make the comparison of data from different sources
very difficult.
An extensive review of the relationship between
SOM and aggregation by
Tisdall and Oades (1982)
made a number of points, which reflect many earlier
findings:
(a) Only part of SOM stabilises aggregates, and this
is generally the ‘fresher’ or ‘active’ material,
with a relatively large concentration of mono- or
poly-saccharides, roots and fungal hyphae.
(b) Above a certain SOM concentration, there is no fur-
ther increase in the stabilising effect, albeit Tisdall
and Oades gave no indication of what this concen-
tration might be.
(c) SOM is not a major binding agent, in the sense
of cementation, presumably because either the
‘binding’ substances are water-soluble or transient,
or both.
(d) Disposition of SOM is more important than the
amount, in that the association of ‘fresh’ compo-
nents with particle surfaces, particularly within ag-
gregates (
Puget et al., 1995
) is the crucial factor.
(e) Particle arrangement is more important in some
soils than is the amount of SOM. For example,
Loveland and Clayden (1987)
showed that close
packing of particles in the size range 31–125
m,
in the presence of
<
0.5% total SOM, could give
rise to a densipan (
D
b
=
1
.
973 Mg m
−
3
) that was
completely free of any form of cementation.
The authors also noted wide variations in the type
and content of SOM between aggregates of differ-
ent sizes, and that aggregate stability can vary widely
within the same soil, depending in part when SOM
was added, what type of SOM it was, and when the
aggregates were subsequently taken for examination.
These factors are part of the difficulty in agreeing on
the role of SOM in aggregate stability, especially in the
derivation of equations of state from limited data. In
general,
Tisdall and Oades (1982)
ranked SOM com-
ponents in terms of the time-scale over which they
exert an influence on aggregate stability and/or forma-
tion:
Page 9
P. Loveland, J. Webb / Soil & Tillage Research 70 (2003) 1–18
9
Fig. 1. Relationship between SOC content (%) and mass of WSAs (w/w) (after
Stengel et al., 1984
).
(1) glucose-like compounds (‘transient mono- and
poly-saccharides’): act strongly after 2–3 weeks,
and decline over the next 4–6 months;
(2) cellulose achieves its maximum effect after 6–9
months, but is never as effective as transient mono-
and poly-saccharides;
(3) ryegrass residues increase in effect over ca. 3
months, persist for another 4–6 months, then de-
cline over the next 3–4 months.
Thus, the most important components of SOM in
soil aggregation exert their effect over no more than
ca. 1 year, which matches the observation that aggre-
gate stability is greatest under grass (near continual
production of such components), but decreases rapidly
under arable cultivation. However,
Tisdall and Oades
(1982)
did not give any threshold values above or be-
low which soil properties were more or less dependent
on a particular level of SOM.
Greenland et al. (1975)
tested the aggregate stability
of 180 samples of English and Welsh soils using the
Emerson crumb test. Although many factors were ex-
amined, the general ‘rule-of-thumb’ conclusion was:
•
soils
<
2% organic carbon: unstable,
•
soils 2–2.5% organic carbon: moderately stable,
•
soils >2.5% organic carbon: very stable.
Stengel et al. (1984)
examined the relationship
between SOC and water stable aggregates (WSAs)
in Ashley (Endostagnic Luvisol), Hamble (Haplic
Luvisol), Wothersome (now Tickenham) (Chromi-
Endostagnic Luvisol), Fincham (now Maxted) (Chro-
mo-Profundic Luvisol), Hanslope (Calcari-Endosta-
gnic
Cambisol),
Wicken
(Calcari-Endostagnic
Cambisol), Denchworth (Eutri-Stagnic Cambisol), La-
wford (Eutri-Stagnic Cambisol), Newchurch (Calcari-
Fluvic Gleysol), Newport (Eutri-Lamellic Arenosol)
and Andover (Calcaric Leptosol) soils (all soil classi-
fications according to
FAO, 1998
), and found (
Fig. 1
)
that:
WSA%
=
11
.
57 SOC%
+
12
.
75
(R
2
=
0
.
61
, P <
0
.
001
)
(1)
Chaney and Swift (1984)
, in a study of more than 120
UK soils, found that total SOC could help to identify
soils at risk of structural degradation, although since
the relation was linear they again gave no threshold
values. Their findings are described by (note the switch
from SOC to SOM):
mean weight diameter
(
MWD
)
=
(
SOM
+
24
)
×
31
(P <
0
.
01
)
(2)
Ekwue (1990)
found a linear relation between SOC
and the amount of WSA in sandy soils under grass at
Woburn:
WSA
(
%
>
0
.
5 mm
)
=
3
.
32 SOC%
−
1
.
44
(n
=
14
, R
2
=
0
.
87
, P <
0
.
001
)
(3)
Page 10
10
P. Loveland, J. Webb / Soil & Tillage Research 70 (2003) 1–18
King and Evans (1989)
showed a linear relationship
graphically between the mass of WSA and SOM for
a range of restored and unmined (sic) soils in north-
ern England. From the graph it can be estimated that
an increase of ca. 1% SOM gives an increase of ca.
8% by mass of WSA, irrespective of land use. In con-
trast to the earlier work (
Carter, 1992
),
Carter et al.
(1994)
found no correlation between amounts of SOM,
either on a whole soil basis or fractionated, and ag-
gregate stability in a range of soils under long-term
cool-temperate Fescue grass in Canada.
By examining soil data from long-term experi-
ments at Woburn, Saxmundham and Rothamsted,
Rose (1990)
found that FYM increased AWC and
aggregate stability by up to 20% in all soils, while
D
b
and macroporosity decreased with increasing use
of FYM on sandy soils, but changed relatively little
on other soils (mostly clay loams and silt loams).
Change in aggregate stability was inferred from
changes in flow rates of water through columns of
dry aggregates, packed to maximum density by gen-
tle shaking, these changes being interpreted as re-
ductions in both intra- and inter-aggregate porosity,
and hence in the stability of the aggregates them-
selves.
On previous evidence, green manuring should im-
prove soil structural properties because of the greater
proportion of ‘fresh’ OM returned to the soil and root
activity of the green manure. However,
Macrae and
Fig. 2. Relationship between SOC content (%) and MWD of aggregates for air-dried (ad) and field moist (fm) samples of some New
Zealand soils (derived from
Haynes and Swift, 1990
).
Mehuys (1987)
were unable to show improvements
in aggregate stability in an 8 treatment
×
3-year ex-
periment involving clover (Trifolium)/vetch (Vicia)
intercropped with maize (Z. mays) on a clay soil
and a gravelly sandy loam in Canada, although D
b
decreased in the green-manured plots
(P <
0
.
05
)
.
Aggregate stability was determined as the weighted
proportion of wet-sieved aggregates, divided by the
weighted proportion of dry-sieved aggregates, in
the sieve aperture range 8 to
<
0.25 mm, in steps
of 8–4.75, 4.75–2, 2–1, 1–0.25 and
<
0.25 mm. The
weighting factors were the same for each sieve aper-
ture for both wet and dry sieving. In contrast to many
of the above studies, however,
Perfect and Kay (1990)
found no relationship between aggregate stability
and SOC content in a Canadian silt loam subject to
a number of land uses. As happens so often in the
studies reviewed here, this conflicts directly with the
work of
Haynes and Swift (1990)
and
Swift (1990)
,
who found that land use and soil water content in-
fluenced aggregate stability more than total SOC
content or poly-saccharide content in New Zealand
silty and silty clay soils of the same mineralogy, in
the sequence: arable
<
pasture (
<
2 years
) <
pasture
(>2 years).
There was considerably more scatter in the data
obtained from air-dried samples, than those from
field-moist samples of the same soils (
Fig. 2
), which
reinforces the message that the hydrologic history of
Page 11
P. Loveland, J. Webb / Soil & Tillage Research 70 (2003) 1–18
11
aggregates is a major factor in their behaviour. How-
ever,
Haynes et al. (1991)
went on to show that in a
sandy loam, silt loam and a clay loam in New Zealand,
the best predictor of MWD of aggregates (determined
as the sum of the fractions of initially air-dried,
4–2 mm aggregates remaining on sieves of aperture 2,
1 and 0.5 mm, after 15 min of sieving under water at
an oscillation rate of 2.5 cycles min
−
1
with an ampli-
tude of 3.5 cm) was the length of time under grass and
the amount of hot-water-soluble poly-saccharide. This
was especially true in soils under pasture for more
than 4 years. The data showed considerable scatter, but
no inflections suggestive of thresholds in SOC values.
The greatest amount of variance in the MWD mea-
surements were explained by the following (across all
treatments, but based on the behaviour of air-dried ag-
gregates):
sandy loam :
MWD
=
0
.
60 SOC%
+
0
.
65
(4)
silt loam :
MWD
=
1
.
09 SOC%
−
0
.
86
(5)
clay loam :
MWD
=
0
.
62 SOC%
+
0
.
27
(6)
Saccharides (arabinose, fructose, galactose, glucose,
mannose, ribitol, xylose and ‘total’) were also re-
ported to be the most important factor in improve-
ments in aggregate stability compared with organic
carbon concentration, D
b
and respiration rate, in an
irrigated coarse loamy soil receiving 25 Mg ha
−
1
per year of organic amendments (below), in each
of 3 years (
Martens and Frankenberger, 1992
). The
amendments were most effective in the order:
barley straw
>
poultry manure
>
sewage sludge
>
alfalfa
Statistical analysis of the data, however, gave weak
correlations. Fructose (
r
=
0
.
42
, P <
0
.
001) was
the single most effective variable involved in aggre-
gate stabilisation in year 1, with ‘total’ saccharides
(
r
=
0
.
39
, P <
0
.
001) the most effective in year 2.
None of the variables showed significant correlation
with aggregate stability in year 3. The effect of SOC
from different sources on aggregate stability was ex-
amined by
Puget et al. (1995)
for a silty soil under
conventional and zero tillage (ZT) in the Paris Basin.
Young C
4
-carbon from
<
6 to 23 years in age, and lo-
cated in the aggregates rather than dispersed through
the soil, was responsible for most of the stability of
soil aggregates >1 mm in size. The SOC contents of
this soil ranged from ca. 1 to 2.5%. Similarly, in
long-term (38 years) experiments at Ultuna, Sweden,
Gerzabek et al. (1995)
reported that SOM that en-
couraged greater production of soil microbial biomass
(SMB), e.g. ‘green manuring’, FYMs gave greater ag-
gregate stability. Thus, fallow plots had the least ag-
gregate stability, and plots receiving FYM, the greatest
(cf.
Rose, 1990
, above). Peat added to soils gave little
increase in SMB, because it contained little ‘young’
or ‘active’ SOC to act as a microbial substrate in the
short-term, and thus failed to improve aggregate sta-
bility. This is paralleled by the work of
Friedel et al.
(1996)
, who also found that the amount of young ‘mi-
crobially active’ SOM was closely related to WSA of
<
2 mm air-dry aggregates (
R
2
=
0
.
94).
Douglas and Goss (1982)
investigated the effect
of different cultivation practices, over a 5-year pe-
riod, on the physical properties of Hamble, Andover,
Lawford and Denchworth soils (classified above) in
southern England. Correlations of SOC with aggre-
gate stability for individual treatments were generally
significant (
R
2
∼
0
.
6–0
.
7,
P <
0
.
001). Aggregate
stability was determined as the turbidity of a suspen-
sion produced by shaking 2 g of air-dry aggregates
(2.8–1.4 mm diameter) with 85 ml of deionised water
(in a stoppered 100 ml tube) for 45 min at 16 rpm at
a temperature of 2
◦
C, the measurement of turbidity
being made immediately upon cessation of shaking.
Pooling of data from all treatments improved the cor-
relations for the Lawford, Hamble and Denchworth
soils to
R
2
=
0
.
727, 0.858 and 0.945 (all
P <
0
.
001),
respectively. Minimum cultivation tended to lead
to greater aggregate stability compared with deep
ploughing, but the most stable aggregates were found
under grass, which is consistent with other work.
Differences between the cultivation treatments of-
ten did not become significant until the fifth year of
treatment. From the graphs presented, there are no
obvious inflection points to indicate a marked change
in behaviour above or below a particular value of
SOC.
A number of studies has summarised the effects of
tillage practices (via changes in SOC) on soil aggre-
gate properties. For example, the MWD of WSAs (2–1
and 9–4.75 mm) in the 0–5 cm layer of a number of
Canadian soils under a range of tillage was found to
be linearly related to SOC by the following expression
Page 12
12
P. Loveland, J. Webb / Soil & Tillage Research 70 (2003) 1–18
(
Carter, 1992
):
MWD%
=
127
(
SOC%
)
−
63
.
4
(R
2
=
0
.
94
, P <
0
.
001
)
(7)
The maximum ‘value’ of soil structural stability, as
expressed by MWD, was found at 4.5% SOC content.
Seasonal variation in aggregate stability was stud-
ied by
Chan et al. (1993)
in Australia, who compared
ZT and conventional tillage (CT) for wheat, with and
without a lupin (Lupinus spp.) break-crop. Lupins im-
proved aggregate stability under CT, but made no dif-
ference under ZT. There was no statistically significant
difference between ZT and CT without L (
n
=
6):
percentage aggregates
(
50 mm
,
CT
+
L
)
=
0
.
376
+
0
.
163
(
SOC%
) (R
2
=
0
.
97
, P <
0
.
1
)
(8)
ditto
(
250 mm
,
CT
+
L
)
= −
0
.
107
+
0
.
303
(
SOC%
) (R
2
=
0
.
95
, P <
0
.
1
)
(9)
Also in Canada, ZT increased the amount of SOC
sequestered in aggregates >0.25 mm in the upper
125 mm of a Gray Luvisol (clay content
=
18%) from
0.16 to 0.34 kg m
−
2
over a period of up to 16 years, as
compared to the same soil under CT (
Franzluebbers
and Arshad, 1996
). Unusually,
Jastrow (1996)
found a
power function relating the length of time under grass
to the amount of SOC in the aggregates themselves,
as expressed by:
aggregate SOC%
=
96
.
6
(
1
−
0
.
637
−
0
.
012 year since cultivation
)
(10)
Such power functions relating aggregate properties to
other measured soil properties are reported rarely in
the literature, most such relationships being reported
as linear.
Rasiah (1994)
and
Rasiah and Kay (1994)
attempted to calculate the half-life for water-stable ag-
gregates in soils under forage, and arrived at 4.5 years
for a clay loam, and 7.75 years for a sandy loam.
However, they gave no data to show the uncertainty
of these modelled results. More recently,
Boix-Fayos
et al. (2001)
found that the relationship between SOM
and aggregate size, in some soils in Spain, differed
depending on the size of the aggregates. Thus, SOM
was negatively correlated to the stability of aggregates
>2 mm, but positively so for aggregates
<
1 mm. They
related this difference to the relative contribution of
earthworm casts to the aggregates, which implies a
strong influence of ‘recent’ SOM.
Few people consider the depth of sampling in stud-
ies of soil aggregate stability.
Unger (1995)
made the
point that many studies on the interaction between
SOM and aggregate stability could be flawed, because
too great a depth of soil was examined. In a study
of ridge-tilling, he found that most benefit from SOM
was confined to the top 4 cm of the soil. Sampling
from greater depths obscured this benefit.
4.2. SOC, D
b
, rheology and shrinkage
In his study of UK soils,
Williams (1971)
found that
D
b
(
arable
)
=
1
.
42
−
0
.
78 total N%
(R
2
=
0
.
38
)
(11)
D
b
(
grass
)
=
1
.
37
−
0
.
76 SOC%
(R
2
=
0
.
69
)
(12)
Graphs presented by
Williams (1971)
showed an in-
flection at ca. 2% SOC, below which little change in
D
b
occurs (note, however, that the number of points
on this graph is remarkably small for a dataset said
to be drawn from 189 soils).
Hamblin and Davies
(1977)
investigated the influence of SOC on the liquid
and plastic limits, air-filled porosity
(
pF
<
5
.
0
)
, pore
size distribution (pF
=
1
.
0–5.0
3
), shrinkage (assumed
to be linear) and D
b
of silty soil series in east-
ern England, namely Beccles—Albi-Luvic Planosol,
Newchurch—Calcari-Fluvic Gleysol, Agney—Eutri-
Fluvic Gleysol, Romney—Endogleyi-Fluvic Cam-
bisol), and Sheringham—Eutric Cambisol (
FAO,
1998
). Although they present their findings in terms
of ‘high organic’ and ‘low organic’ soils, they do not
give any numerical definition of ‘high’ and ‘low’. Val-
ues of D
b
between ca. pF2 and pF5 are consistently
0.2 g cm
−
3
greater in the ‘low organic’ soils than in
the ‘high organic’ soils, as described by:
low organic soils :
D
b
(
g cm
−
3
)
=
1
.
31
+
0
.
12pF
(13)
3
The numerical relationships given by
Hamblin and Davies
(1977)
are expressed in terms of pF and are quoted verbatim. The
relationships are linear in terms of this power function. In the
absence of their raw data, we are unable to make an exact trans-
lation of the relationship in terms of, for example, the commonly
used SI unit, the Pascal (Pa). For purposes of comparison, pF5 is
9
.
8
×
10
7
Pa and pF2 is 9
.
8
×
10
4
Pa.
Page 13
P. Loveland, J. Webb / Soil & Tillage Research 70 (2003) 1–18
13
high organic soils :
D
b
(
g cm
−
3
)
=
1
.
07
+
0
.
13pF
(14)
The correlation between D
b
and total SOC at pF 2.5
was highly significant (
R
2
=
0
.
94,
P <
0
.
001), as was
the negative correlation with maximum compaction
density (
R
2
= −
0
.
93,
P <
0
.
001), the latter being
measured using a Dietert apparatus (
HMSO, 1952
)—
a method similar in principle to the widely known
‘Proctor’ compaction test. Shrinkage was ca. 5% less
at pF2 to
<
1% less at pF4 in the low organic soils,
and was described by the following equations:
low organic soils :
shrinkage
=
27
.
63
−
5
.
04pF
(15)
high organic soils :
shrinkage
=
31
.
8
−
5
.
88pF
(16)
Shiel and Rimmer (1984)
also carried out a study of
the relationship between SOC and D
b
at Cockle Park
(northern England), and gave the relationship:
D
b
(
g cm
−
3
)
=
1
.
62
×
0
.
134 SOC%
(R
2
=
0
.
99
)
(17)
In a rather similar investigation to that of
Douglas
and Goss (1982)
,
Carter (1987)
found a distinctly
non-linear relationship in a study of the influence of
tillage practices on the relationship between D
b
and
SOC content, and not the straight line shown or im-
plied by most other workers. There are no inflections
in this curve over the range of SOC values shown (ca.
0.3–2.1%)
D
b max
=
1
.
62 SOC%
−
0
.
079
(R
2
=
0
.
90
, n
=
16
)
(18)
Ekwue (1990)
gave the following relationship for D
b
in sandy soils under grass from Woburn:
D
b
=
1
.
61
−
0
.
09 SOC%
(R
2
= −
0
.
977
, P <
0
.
001
, n
=
29
)
(19)
Cannell et al. (1994)
also found a linear relationship
between D
b
and SOM in soils in a cultivation experi-
ment extending over 18 years of different tillage prac-
tices:
D
b
= −
0
.
102
(
OM%
)
+
2
.
07
(R
2
=
0
.
93
, P <
0
.
001
)
(20)
4.3. SOC, soil porosity and water retention
properties
Hamblin and Davies (1977)
found that all the ‘low
organic’ soils had lower total porosities at all suctions
than the ‘high organic’ soils. From the graphs pre-
sented in their papers, these differences were
<
1% at
pF1 to almost 10% at pF4, although most were
<
5%.
Water content at pF2 correlated strongly with total
SOC (
R
2
=
0
.
91,
P <
0
.
001). Some of these differ-
ences in behaviour were attributed by
Hamblin (1977)
to the fact that the high SOC soils contained greater
amounts of humified SOM in the 2–50
m fractions
than the low SOC soils, whilst the latter contained
more humified SOM in the
<
2
m fractions.
There have been numerous other attempts to relate
soil water retention properties to SOC or SOM.
Hollis
et al. (1977)
examined the relationships between SOC
and both water held at
−
5 kPa
(
−
5
Θ
vf
)
, i.e. water ex-
tracted as the soil dries and available water (A
vf
) (de-
fined as the difference between the water content of
the soil at a matric potential of 5 and at 1500 kPa)
within 77 soil profiles from the west midlands of Eng-
land. Water retention properties are expressed on a
stone-free, volumetric basis, i.e. all stones >2 mm are
sieved from the soil after the water retention prop-
erties are measured. The curvilinear relationship be-
tween SOC and
−
5
Θ
vf
in the surface horizons was
described by
−
5
Θ
vf
=
23
.
88
+
7
.
85 SOC
−
0
.
43 SOC
2
(
% variance explained
=
73
.
5
, P <
0
.
001
)
(21)
whereas the relationship with A
vf
was linear in both
surface and subsoils, and best described for the com-
bined horizon data by
A
vf
=
15
.
6
+
2
.
14
(
SOC
)
(
% variance explained
=
48
.
9
, P <
0
.
001
)
(22)
The plotted curves are smooth with no apparent clus-
tering of the data. The authors make the point that
where SOC is the same in different soils, then dif-
ferences in
−
5
Θ
vf
and A
vf
can be due to different
amounts of clay and silt. A further study of UK Soil
Survey data (Bellamy and Loveland, unpublished)
Page 14
14
P. Loveland, J. Webb / Soil & Tillage Research 70 (2003) 1–18
Table 2
Contribution of SOC to soil water content at different tensions (% variance accounted for) for a range of UK soils
Factor
√
%Clay
+
Silt%
+
ln(SOC%)
SOC contribution
n
Topsoils (arable)
θ
(
−
5 kPa)
62.8
66.2
76.4
10.2
248
θ
(
−
10 kPa)
67.9
71.7
78.8
7.1
222
θ
(
−
40 kPa)
70.5
74.2
79.4
5.2
222
θ
(
−
200 kPa)
72.8
74.9
79.7
4.8
208
θ
(
−
1500 kPa)
79.0
79.0
82.1
3.1
248
Topsoils (permanent grassland)
θ
(
−
5 kPa)
38.7
42.3
55.7
13.4
238
θ
(
−
10 kPa)
46.2
51.8
61.6
9.8
227
θ
(
−
40 kPa)
56.7
64.1
68.8
4.7
227
θ
(
−
200 kPa)
54.6
59.3
61.8
2.5
178
θ
(
−
1500 kPa)
64.7
65.5
67.8
2.3
238
showed that SOC contributes no more than ca. 13%
to the variance in soil water content, and often much
less (
Table 2
).
Carter (1992)
found a linear relationship between
Θ
vf
(
−
1500 kPa) and SOC in the soil layer 0–5 cm
depth under a range of tillage practices in Canada
Θ
vf
(
−
1500 kPa
)
=
7
.
2
+
2
.
3
(
SOC%
)
(R
2
=
0
.
97
, P <
0
.
001
)
(23)
4.4. Other soil physical properties
Kemper et al. (1987)
found that, in a series of recon-
stituted soils (5–40% clay;
<
0.6–1.2% SOC), recov-
ery of cohesion was ca. 100 times faster in soils with
>35% clay and
<
0.6% SOC, than in the other soils. It
was assumed that this reflected the ability of clay par-
ticles to form stable inter-particle bonds more quickly
in the absence of SOC. In sandy soils in North Ger-
many, cohesion and the angle of friction were found
to be controlled by both SOM concentration and water
held at suctions
<
−
50 kPa, but only where SOM >5%
(
Bachmann and Zhang, 1991
). In sandy soils from
Woburn,
Ekwue (1990)
found that shear strength was
related to SOC by
shear strength
=
20
.
61
+
0
.
67 SOC%
(n
=
29
, R
2
=
0
.
52
, P <
0
.
001
)
(24)
Auerswald (1995)
found a linear relationship between
‘percolation stability’ (PS) (see also
Greenland et al.,
1962
), pH and SOM content in aggregates 1–2 mm
diameter from arable topsoils
PS
=
(
14
−
pH
)
×
SOM%
(R
2
=
0
.
72
, P <
0
.
001
, n
=
113
)
(25)
where PS was determined from the incremental change
in flow rate of demineralised water, at a constant head
of 20 kPa, as it passed through 10 g of close-packed (cf.
Rose, 1990
), air-dried aggregates, 2–1 mm diameter,
over a period of 10 min.
In conclusion, it is also noteworthy, from our ex-
amination of the literature, how little standardisation
there is of procedures for the determination of aggre-
gate stability. This makes inter-comparison of research
findings considerably more difficult than it might oth-
erwise be.
4.5. Erosion
Soil erosion is one aspect of soil physical be-
haviour in which SOC content is regarded as a factor.
Data for the UK appear to be lacking, but
Benito and
Diaz-Fierros (1992)
presented data from Spain, from
which
Fig. 3
is derived. The relationship for soils un-
der cereals is, effectively, non-existent, whilst that for
soils under pasture is more definite. From the graph,
it would seem that soil loss decreases markedly above
an SOC content of about 3%.
Under Australian conditions,
Malinda (1995)
found
that soil loss from 10-year plots under wheat, was
strongly controlled by SOC content, with a steep
Page 15
P. Loveland, J. Webb / Soil & Tillage Research 70 (2003) 1–18
15
Fig. 3. Relationship between SOC content (%) and soil loss (g m
−
2
per month) under different land-uses in Spain (derived from
Benito
and Diaz-Fierros, 1992
).
increase in amounts of soil loss at SOC contents
<
1.6%
soil loss
(
t ha
−
1
per year
)
=
31
+
16
(
SOC%
)
−
45
(
√
%SOC
) (R
2
=
0
.
76
)
(26)
5. Conclusions
The greatly increased crop yields of the last half
century have often been obtained despite decrease in
SOC, often to levels below those posed as critical.
The literature suggests that where reductions in SOC
lead to reductions in yield potential these are small,
and have not lead to a crisis in crop production. Some
studies, usually on eroded soils, indicate that as SOC
decreases it becomes increasingly difficult to obtain
yields equivalent to those on un-degraded soils by
the addition of fertiliser alone, albeit yields can be
maintained at ca. 90% of potential. Although many
studies have demonstrated numerical relationships
between SOC or SOM and various soil properties,
firm evidence of a threshold above or below which
the contribution of carbon increases or decreases
significantly is rare. There is some suggestion from
cropping data that such a threshold, in relation to
N-supply from plant residues, might be at ca. 1%
SOC. At this point, decomposition of SOM is in
equilibrium with OM additions from plant residues,
and thus SOC remains almost constant. Under these
conditions, there is some evidence that cereal yields
will be about 90–95% of the maximum, assuming
adequate supplies of inorganic fertilisers.
Similarly, widespread evidence for a threshold of
SOC in relation to soil physical properties, below
which catastrophic failure of, for example, soil ag-
gregation will occur, is lacking. There is, however,
a growing body of evidence that ‘fresh’ or ‘active’
plant residues are important in determining soil ag-
gregation and related properties, and the proportion
of these residues in the total SOC pool is more im-
portant than the size of the pool itself. We believe
that further investigation of the relationship between
‘active’ and ‘total’ soil carbon and soil properties will
enable a better understanding of soil behaviour, an
improvement of our understanding of soil quality and
related issues, and the quality of the advice provided
about them. A single critical threshold value for soil
carbon content in temperate soils cannot be supported
from the evidence available, which is also inadequate
for the establishment of different critical SOC val-
ues for different soils under different land uses. The
evidence available suggest that there might be an op-
timum or desirable range of SOM across a range of
soil types, but again, we find the evidence equivocal
at best. Considerably more quantitative investigation
would be required to establish this clearly. The debate
on SOM will continue, if for no other reason than al-
most everyone sees it as a keystone indicator for soil
Page 16
16
P. Loveland, J. Webb / Soil & Tillage Research 70 (2003) 1–18
quality or soil ‘health’. However, the nature of this
indicator (total SOM or SOC, ‘active’ SOM or some
fraction of it, etc.) will need further investigation if
it is to be widely accepted in the long term, with the
parallel implications for design, implementation and
costs of soil monitoring frameworks.
Acknowledgements
This review derives from Contract SP0306: ‘Crit-
ical Levels of Soil Organic Matter’, financed by the
Ministry of Agriculture, Fisheries and Food, London,
UK.
References
Addiscott, T.M., 1970. A short note resolving soil cation exchange
capacity into ‘mineral’ and ‘organic’ fractions. J. Agric. Sci.
75, 365–367.
Allison, F.E., 1973. Soil Organic Matter and its Role in Crop
Production. Elsevier, New York.
Anon., 1994. Fertiliser Recommendations for Agricultural and
Horticultural Crops. MAFF Reference Book 209. HMSO,
London.
Arrouays, D.O., Pelissier, P., 1994. Changes in carbon storage in
temperate loamy humic soils after forest clearing and continuous
corn cropping in France. Plant Soil 160, 215–223.
Arrouays, D., Deslais, W., Badeau, V., 2001. The carbon content
of topsoil and its geographical distribution in France. Soil Use
Manage 17, 7–11.
Auerswald, K., 1995. Percolation stability of aggregates from
arable topsoils. Soil Sci. 159, 142–148.
Bachmann, J., Zhang, H., 1991. The stability of sandy soils and
its dependence on humification degree and content of organic
matter. Zeitschr. Pflanzener. Boden. 154, 47–52.
Battiston, L.A., Miller, M.H., Shelton, J.J., 1987. Soil erosion and
corn yield in Ontario. 1. Field Evaluation. Can. J. Soil Sci. 67,
731–745.
Bauer, A., Black, A.L., 1992. Organic carbon effects on available
water capacity of three soil textural groups. Soil Sci. Soc. Am.
Proc. 56, 248–254.
Bauer, A., Black, A.L., 1994. Quantification of the effect of soil
organic matter content on soil productivity. Soil Sci. Soc. Am.
J. 58, 185–193.
Beare, M.H., Hu, S., Coleman, D.C., Hendrix, P.F., 1997.
Influences of mycelial fungi on soil aggregation and organic
matter storage in conventional and no-tillage soils. Appl. Soil
Ecol. 5, 211–219.
Beck, A.J., Jones, K.C., Hayes, M.H.B., Mingelgrin, U. (Eds.),
1993. Organic Substances in Soil and Water. Royal Society of
Chemistry, Cambridge, UK.
Bell, M.J., Moody, P.W., Yo, S.A., Connolly, R.D., 1999. Using
active fractions of soil organic matter as indicators of the
sustainability of Ferralsol farming systems. Aust. J. Soil Res.
37, 279–287.
Benito, E., Diaz-Fierros, F., 1992. Effects of cropping on the
structural stability of soils rich in organic matter. Soil Till. Res.
23, 153–161.
Boix-Fayos, C., Calvo-cases, A., Imeson, A., Soriano-Soto, M.D.,
2001. Influence of soil properties on the aggregation of some
Mediterranean soils and the use of aggregate size and stability
as land degradation indicators. Catena 44, 47–67.
Cannell, R.Q., Hawes, J.D., Jensen, H.E., 1994. Trends in tillage
practices in relation to sustainable crop production with special
reference to temperate climates. Soil Till. Res. 30, 245–282.
Carter, M.R., 1987. Physical properties of some Prince Edward
Island soils in relation to their tillage requirement and suitability
for direct drilling. Can. J. Soil Sci. 67, 473–487.
Carter, M.R., 1992. Influence of reduced tillage systems on organic
matter, microbial biomass, macro-aggregate distribution and
structural stability of the surface soil in a humid climate. Soil
Till. Res. 23, 361–372.
Carter, M.R., Stewart, B.A. (Eds.), 1996. Structure and Organic
Matter Storage in Agricultural Soils. CRC Press, Boca Raton.
Carter, M.R., Angers, D.A., Kunelius, H.T., 1994. Soil structural
form and stability, and organic matter under cool-season
perennial grasses. Soil Sci. Soc. Am. J. 58, 1194–1199.
Chan, K.Y., Heenan, D.P., Ashley, R., 1993. Seasonal changes
in surface aggregate stability under different tillage and crops.
Soil Till. Res. 28, 301–314.
Chaney, K., Swift, R.S., 1984. The influence of organic matter
on aggregate stability in some British soils. J. Soil Sci. 35,
223–230.
Chesters, G., Attoe, O.J., Allen, O.N., 1957. Soil aggregation in
relation to various soil constituents. Soil Sci. Soc. Am. Proc.
21, 272–277.
Clayden, B., Hollis, J.M., 1984. Criteria for differentiating soil
series. Soil Surv. Tech. Monogr. no. 17.
Curtin, D., Smillie, G.W., 1976. Estimation of components of soil
cation exchange capacity from measurements of specific surface
and organic matter. Soil Sci. Soc. Am. J. 40, 461–462.
Davidson, D.A., 2000. Soil quality assessment: recent advances
and controversies. Prog. Environ. Sci. 2, 342–350.
Department of the Environment, 1997. Sustainable Use of Soil:
Government Response to 19th Report of RCEP. HMSO London.
Douglas, J.T., Goss, M.J., 1982. Stability and organic matter
content of surface soil aggregates under different methods of
cultivation and grassland. Soil Till. Res. 2, 155–175.
Ekwue, E.I., 1990. Organic matter effects on soil strength
properties. Soil Till. Res. 16, 289–297.
Fagbenro, J.A., Agboola, A.A., 1999. Effect of aerial oxidation
pre-treatment on composition and effectiveness of a Nigerian
lignite sample as soil organic matter amendment. Comm. Soil
Plant Anal. 30, 2331–2362.
FAO, 1998. World reference base for soil resources. World Soil
Resources Report No. 84. FAO, Rome, 88 pp.
Figliola, A., Izza, C., Biondi, F.A., Indiati, R., Mangione, D.,
1993. Influence of organic matter on the behaviour of copper
Page 17
P. Loveland, J. Webb / Soil & Tillage Research 70 (2003) 1–18
17
and zinc in the soil and their availability for crop uptake. In:
El estudio del suelo y de su degradación en relación con la
desertificación. Actas del 12 Congresso Latinoamericans de la
Cienca del Suelo, Sevilla, pp. 668–675.
Franzluebbers, A.J., Arshad, M.A., 1996. Water-stable aggregation
and organic matter in four soils under conventional and zero
tillage. Can. J. Soil Sci. 76, 387–393.
Friedel, J.K., Munch, J.C., Fischer, W.R., 1996. Soil microbial
properties and the assessment of available soil organic matter
in a Haplic Luvisol after several years of different cultivation
and crop rotation. Soil Biol. Biochem. 28, 479–488.
Gerzabek, M.H., Kirchmann, H., Pichlmayer, F., 1995. Response
of soil aggregate stability to manure amendments in the Ultuna
long-term soil organic matter experiment. Zeitschr. Pflanzener.
Boden. 158, 257–260.
Glendining, M.J., Powlson, D.S., Poulton, P.R., Bradbury, N.J.,
Palazzo, D., Li, X., 1996. The effects of long-term applications
of inorganic nitrogen fertiliser on soil nitrogen in the broadbalk
wheat experiment. J. Agric. Sci. Camb. 127, 347–363.
Grace, P.R., Oades, J.M., Keith, H., Hancock, T.W., 1995. Trends
in wheat yields and soil organic carbon in the permanent
rotation trial at the Waite Agricultural Research Institute, South
Australia. Aust. J. Exp. Agric. 35, 857–864.
Greenland, D.J., Lindstrom, G.R., Quirk, J.P., 1962. Organic
materials which stabilise natural soil aggregates. Soil Sci. Soc.
Am. Proc. 26, 366–371.
Greenland, D.J., Rimmer, D., Payne, D., 1975. Determination of
the structural stability class of English and Welsh soils, using
a water coherence test. J. Soil Sci. 26, 294–303.
Hamblin, A.P., 1977. Structural factors of aggregation in some
east Anglian soils. J. Soil Sci. 28, 23–28.
Hamblin, A.P., Davies, D.B., 1977. Influence of organic matter on
the physical properties of some east Anglian soils of high silt
content. J. Soil Sci. 28, 11–22.
Hayes, M.H.B., Wilson, W.S. (Eds.), 1997. Humic Substances,
Peats and Sludges. The Royal Society of Chemistry, Cambridge.
Haynes, R.J., Naidu, R., 1998. Influence of lime, fertiliser and
manure applications on soil organic matter content and soil
physical conditions: a review. Nutr. Cycl. Agroecosyst. 51, 123–
137.
Haynes, R.J., Swift, R.S., 1990. Stability of soil aggregates in
relation to organic constituents and soil water content. J. Soil
Sci. 41, 73–84.
Haynes, R.J., Swift, R.S., Stephen, R.C., 1991. Influence of mixed
cropping rotations (pasture-arable) on organic matter content,
water-stable aggregation and clod porosity in a group of soils.
Soil Till. Res. 19, 77–87.
HMSO, 1952. Soil Mechanics for Road Engineers. Her Majesty’s
Stationery Office, London.
Hollis, J.M., Jones, R.J.A., Palmer, R.C., 1977. The effects of
organic matter and particle size on the water-retention properties
of some soils in the west midlands of England. Geoderma 17,
225–238.
Howard, P.J.A., Howard, D.M., 1990. Use of organic carbon and
loss-on-ignition to estimate soil organic matter in different soil
types and horizons. Biol. Fertil. Soils 9, 306–310.
Janzen, H.H., 1987. Soil organic-matter characteristics after
long-term cropping to various spring wheat rotations. Can. J.
Soil Sci. 67, 845–856.
Jastrow, J.D., 1996. Soil aggregate formation and the accrual of
particulate and mineral-associated organic matter. Soil Biol.
Biochem. 28, 665–676.
Jenkinson, D.S., Rayner, J.H., 1977. The turnover of soil organic
matter in some of the Rothamsted classic experiments. Soil Sci.
123, 298–305.
Johnston, A.E., 1991. Soil fertility and soil organic matter. In:
Wilson, W.S. (Ed.), Advances in Soil Organic Matter Research:
The Impact on Agriculture and the Environment. The Royal
Society of Chemistry Cambridge, pp. 297–314.
Kalembasa, S.J., Jenkinson, D.S., 1973. A comparative study
of titrimetric and gravimetric methods for the determination
of organic carbon in soil. J. Sci. Food Agric. 24, 1085–
1090.
Kemper, W.D., Koch, E.J., 1966. Aggregate stability of soils from
Western United States and Canada. USDA Technical Bulletin
No. 1355, Washington, DC, USA.
Kemper, W.D., Rosenau, R.C., Dexter, A.R., 1987. Cohesion
development in disrupted soils as affected by clay and organic
matter content. Soil Sci. Soc. Am. J. 51, 860–867.
King, J.A., Evans, E.J., 1989. The growth of spring barley related
to soil tilth produced on restored opencast and unmined land.
Soil Till. Res. 14, 115–130.
Körschens, M., Weigel, A., Schulze, E., 1998. Turnover of
soil organic matter (SOM) and long-term balances—tools for
evaluating sustainable productivity of soils. Zeitschr. Pflanzener.
Boden. 161, 409–424.
Lampkin, N., Spedding, C.R.W., 1990. Organic Farming. Farming
Press Books and Videos, UK.
Loveland, P.J., Clayden, B., 1987. A hardpan podzol at Yarner
Wood, Devon. J. Soil Sci. 38, 357–367.
Macrae, R.J., Mehuys, G.R., 1987. Effects of green manuring in
rotation with corn on the physical properties of two Québec
soils. Biol. Agric. Hortic. 4, 257–270.
MAFF, 1998. The Soil Code (Revised 1998): Code of Good
Agricultural Practice for the Protection of Soil. MAFF
Publications, London.
Malinda, D.K., 1995. Factors in conservation farming that reduce
erosion. Aust. J. Expl. Agric. 35, 969–978.
Mamo, M., Molina, J.A.E., Rosen, C.J., Halbach, T.R., 1999.
Nitrogen and carbon mineralization in soil amended with
municipal solid waste compost. Can. J. Soil Sci. 79, 535–
542.
Martel, Y.A., De Kimpe, C.R., Leverdiére, M.R., 1978. Cation
exchange capacity of clay-rich soils in relation to organic matter,
mineral composition and surface area. Soil Sci. Soc. Am. J.
42, 764–767.
Martens, D.A., Frankenberger, W.T., 1992. Modification of
infiltration rates in an organic-amended irrigated soil. Agron.
J. 84, 707–717.
Matus, F.J., Rodr´ıguez, J., 1994. A simple model for estimating the
contribution of nitrogen mineralization to the nitrogen supply
of crops from a stabilised pool of soil organic matter and recent
organic input. Plant Soil 162, 259–271.
Page 18
18
P. Loveland, J. Webb / Soil & Tillage Research 70 (2003) 1–18
McGrath, S.P., Loveland, P.J., 1992. The Soil Geochemical Atlas
for England and Wales. Blackie Academic, Glasgow, 101 pp.
McGrath, S.P., Sanders, J.R., Shalaby, M.H., 1988. The effect of
soil organic matter levels on soil solution concentrations and
extractabilities of manganese, zinc and copper. Geoderma 42,
177–188.
Miller, R.M., Jastrow, J.D., 1990. Hierarchy of root and
mycorrhizal fungal interactions with soil aggregation. Soil Biol.
Biochem. 22, 579–584.
Perfect, E., Kay, B.D., 1990. Relationship between aggregate
stability and organic components for a silt loam. Can. J. Soil
Sci. 70, 731–735.
Pretty, J., 1998. The Living Land. Earthscan, London.
Puget, P., Chenu, C., Balesdent, J., 1995. Total and young organic
matter distributions in aggregates of silty cultivated soils. Eur.
J. Soil Sci. 46, 449–459.
Rasiah, V., 1994. Equations to predict measures of structural
stability at minimum strength. Soil Sci. 158, 170–173.
Rasiah, V., Kay, B.D., 1994. Characterizing changes in aggregate
stability subsequent to introduction of forages. Soil Sci. Soc.
Am. J. 58, 935–942.
Robbins, C.W., Mackey, B.E., Freeborn, L.L., 1997. Improving
exposed subsoils with fertilisers and crop rotations. Soil Sci.
Soc. Am. J. 61, 1221–1225.
Rose, D.A., 1990. The effect of long-continued manuring on some
physical properties of soils. In: Wilson, W.S. (Ed.), Advances in
Soil Organic Matter Research: The Impact on Agriculture and
the Environment. The Royal Society of Chemistry, Cambridge,
UK.
Shepherd, M.A., Stockdale, E.A., Powlson, D.S., Jarvis, S.C.,
1996. The influence of organic nitrogen mineralization on
the management of agricultural systems in the UK. Soil Use
Manage 12, 76–85.
Shiel, R.S., Rimmer, D.L., 1984. Changes in soil structure and
biological activity on some meadow hay plots at Cockle Park,
Northumberland. Plant Soil 76, 349–356.
Smith, P., Smith, J.U., Powlson, D.P., 1996. Soil organic matter
network (SOMNET): 1996 model and experimental metadata.
Global Change and Terrestrial Ecosystems Report No. 7. GCTE
Focus 3 Office, Wallingford.
Stengel, P., Douglas, J.T., Guérif, J., Goss, M.J., Monnier, G.,
Cannell, R.Q., 1984. Factors influencing the variation of some
properties of soils in relation to their suitability for direct
drilling. Soil Till. Res. 4, 35–53.
Swift, R.S., 1990. Effects of humic substances and poly-sac-
charides on soil aggregation. In: Wilson, W.S. (Ed.), Advances
in Soil Organic Matter Research: The Impact on Agriculture
and the Environment. Royal Society of Chemistry Cambridge,
pp. 153–162.
Tagwira, F., Piha, M., Mugwira, L., 1992. Effect of pH, and
phosphorus and organic matter contents on zinc availability and
distribution in two Zimbabwean soils. Comm. Soil Sci. Plant
Anal. 23, 1485–1500.
Tan, K.H., Dowling, P.S., 1984. Effect of organic matter on CEC
due to permanent and variable changes in selected temperate
region soils. Geoderma 32, 89–101.
Tanaka, D.L., Aase, J.K., 1989. Influence of topsoil removal and
fertiliser application on spring wheat yields. Soil Sci. Soc. Am.
J. 53, 228–232.
Thomasson, A.J., Carter, A.D., 1989. Current and future uses
of the UK soil water retention dataset. In: van Genuchten,
M.Th., Leij F.J., Lund L.J. (Eds.), Indirect Methods for
Estimating the Hydraulic Properties of Unsaturated Soils,
Proceedings of an International Workshop, Riverside, CA,
USA, October 11–13. USDA and Department Soil and
Environmental Sciences, Riverside, CA, USA, pp. 355–
358.
Thompson, M.L., Zhang, H., Kazemi, M., Sandor, J., 1989.
Contribution of organic matter to cation exchange capacity and
specific surface area of fractionated soil materials. Soil Sci.
148, 250–257.
Tisdall, J.M., 1994. Possible role of soil microorganisms in
aggregation in soils. Plant Soil 159, 115–121.
Tisdall, J.M., Oades, J.M., 1982. Organic matter and water-stable
aggregates in soils. J. Soil Sci. 33, 141–163.
Unger, P.W., 1995. Organic matter and water-stable aggregate
distribution in ridge-tilled surface soil. Soil Sci. Soc. Am. J.
59, 1141–1145.
Verity, G.E., Anderson, D.W., 1990. Soil erosion effects on soil
quality and yield. Can. J. Soil Sci. 70, 471–484.
Wadman, W.P., De Haan, S., 1997. Decomposition of organic
matter from 36 soils in a long-term pot experiment. Plant Soil
189, 289–301.
Watts, C.W., Whalley, W.R., Longstaff, D.J., White, R.P., Brookes,
P., Whitmore, A.P., 2001. Aggregation of a soil with different
cropping histories following the addition of organic materials.
Soil Use Manage 17, 263–268.
Williams, R.J.B., 1971. Relationships between the composition
of soils and physical measurements made on them. Report of
Rothamsted Experimental Station for 1970, Part 2. Harpenden,
pp. 5–35.
Zhukov, A.I., Sorokina, L.V., Mosaleva, V.V., 1993. Humus and
grain-crop yield on sod-podzolic loamy sand soil. Eurasian Soil
Sci. 25, 82–91.
Zier, N., Schiene, R., Koch, H., Fischer, K., 1999. Agricultural
reclamation of disturbed soils in a lignite mining area using
municipal and coal wastes: the humus situation at the beginning
of reclamation. Plant Soil 213, 241–250.
Document Outline